ETA accuracy benchmark · 🇺🇸 Corpus Christi
How Port Flow's predicted ETA compares to each vessel's own broadcast ETA at Corpus Christi, measured on closed voyages over the last 7 days. We publish the number — good or bad. Lower error is better.
Port Flow predicted ETA (RMSE)
1.28 h
MAE 4.61 h · 31 voyages
Broadcast ETA (RMSE)
63.64 h
crew-declared, 21 voyages (same set)
Our advantage
98.0 %
lower error than the broadcast ETA
Broadcast ETAs more than 7 days from the actual arrival are excluded as AIS sentinel / placeholder values (10 excluded). Both figures are computed on the same remaining voyage set — apples to apples.
Closed voyages
error = predicted/broadcast minus actual arrival (hours)| MMSI | Cargo | Arrival | Our error | Broadcast error |
|---|---|---|---|---|
| 255915876 | product | 07/11, 11:26 PM UTC | -0.68 h | -26.43 h |
| 538004576 | product | 07/11, 02:38 PM UTC | 2.65 h | -44.64 h |
| 352002093 | product | 07/11, 12:32 PM UTC | 0.14 h | -3.04 h |
| 311000937 | product | 07/11, 11:15 AM UTC | 0.40 h | -21.05 h |
| 538003440 | product | 07/11, 06:36 AM UTC | -0.60 h | -53.61 h |
| 255915876 | product | 07/11, 06:09 AM UTC | -0.72 h | -9.16 h |
| 367455770 | chemical | 07/10, 11:09 PM UTC | -0.40 h | -339.16 h |
| 238020000 | product | 07/10, 11:05 PM UTC | 0.04 h | -447.09 h |
| 538012188 | product | 07/10, 09:26 PM UTC | 0.41 h | -89.45 h |
| 538011737 | product | 07/10, 06:15 PM UTC | 3.12 h | -739.26 h |
| 636022580 | product | 07/10, 10:11 AM UTC | -0.53 h | -915.18 h |
| 368004000 | product | 07/10, 03:22 AM UTC | -0.07 h | -134.38 h |
| 538010648 | product | 07/10, 01:10 AM UTC | 0.25 h | -83.17 h |
| 368388590 | product | 07/09, 08:12 PM UTC | -0.02 h | -629.84 h |
| 538003932 | product | 07/09, 04:02 PM UTC | 0.47 h | -94.04 h |
| 636022279 | product | 07/09, 02:54 AM UTC | 2.07 h | -109.92 h |
| 368004000 | product | 07/08, 03:15 PM UTC | -63.34 h | -471.26 h |
| 538006921 | product | 07/08, 12:22 PM UTC | 0.25 h | -30.37 h |
| 352003674 | product | 07/08, 03:35 AM UTC | 0.83 h | -932.59 h |
| 372628000 | product | 07/08, 01:08 AM UTC | -50.49 h | -333.04 h |
| 538003932 | product | 07/07, 07:30 PM UTC | -0.72 h | -49.52 h |
| 636021627 | product | 07/07, 03:47 PM UTC | 3.39 h | -73.79 h |
| 563326000 | product | 07/07, 10:12 AM UTC | 0.82 h | -9.21 h |
| 352003817 | container | 07/07, 12:44 AM UTC | 1.92 h | -65.74 h |
| 538010648 | product | 07/06, 06:21 PM UTC | 2.94 h | -400.37 h |
| 256346000 | product | 07/06, 04:20 PM UTC | 2.13 h | -93.34 h |
| 241593000 | product | 07/06, 03:53 PM UTC | 0.03 h | -46.40 h |
| 538006520 | container | 07/06, 03:28 PM UTC | 0.13 h | -4.47 h |
| 538005462 | product | 07/06, 11:20 AM UTC | 0.47 h | -29.34 h |
| 538006743 | product | 07/06, 05:46 AM UTC | 0.07 h | -8.77 h |
| 538005538 | product | 07/05, 10:28 PM UTC | 2.91 h | -792.47 h |
See it on a live desk
Predicted ETAs, congestion and sanctions screening across 51 ports — try it in one click, no signup.
How it's measured
- Predicted ETA: distance-to-berth over speed-made-good, refreshed on every AIS position, with a seasonal correction from rolling medians.
- Broadcast ETA: the crew-entered ETA in the vessel's AIS message — the de-facto industry default.
- Both are compared to the actual arrival timestamp on the same closed voyages. We report RMSE (penalises large misses) and MAE (typical error).
- Broadcast ETAs more than 7 days off are dropped as AIS sentinel / placeholder values, and our model is scored on that same set — so the comparison is honest, not cherry-picked.
- Coverage is public AIS only; ports with weak coverage show fewer closed voyages. We'd rather show a small honest sample than a big misleading one.